Unverified85% confidenceFactTime unknown
o1系列模型在国际数学奥林匹克预选题上的表现达到了金牌水平
1
Sources
85%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
awesome-LLM-resources：GitHub 8K Star最全大语言模型学习资源库解析
githubWangRongsheng
Related Entities
Related Claims
UnverifiedAlphaGeometry在国际数学奥林匹克历年几何题测试中解决了30道题中的25道，接近金牌选手水平58% similarUnverified在AID基准测试中，o1 high（对应o1 pro）的编程能力得分超过60，优于o3-mini-high，但两者非常接近53% similarUnverified两款模型的最终测试总比分为平局52% similarUnverified在编程基准测试中，Codex-1的表现超越了Claude 3.7和O3 High52% similarUnverified在Pass@4通过率测试中，o1 pro的表现显著优于o150% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/32608API
curl https://kongchang.com/api/v1/knowledge/claims/32608MCP
get_claim(id=32608)